臺灣博碩士論文加值系統

利用外形描述子描述影像，是MPEG-7技術中的影像描述特徵之ㄧ，在我們之前的研究中也發展出一個很好的外形描述方法〝以距離角度分佈為基礎的分層外形描述方法〞。本論文主要是讓我們所研究出的方法，結合以影像內容物件外形區域為主的描述方法 (Region-Based Description)之ART(Angular Radial Transform) ，以及影像內容物件外形輪廓為主的描述方法(Contour-Based Description)之CSS(Curvature Scale-Shape) ，成為一個多描述方法的檢索介面。當您使用我們的檢索系統時，想要檢索一個商標(trademark)或圖示(logo)時，您可以利用已知影像的完整特徵，在資料庫內尋找所想要的影像；對於複雜的圖形，如果無法取得有效的輸入查詢影像，則可於我們所提供的繪圖板上描繪簡單的影像，然後再進行檢索。在資料庫方面，我們也區分為完整且封閉輪廓影像資料庫以及複雜且多區域影像資料庫，讓使用者可於不同的資料庫內做影像檢索。本論文所提供的介面是人性化且容易操作的。

Shape is the most powerful technique in MPEG-7 for retrieval of content information from images. We ever proposed a shape descriptor schema-“Distance and angle-based Layered Shape Feature Extraction Technique”. This issues propose a new image retrieve system which combine three techniques（Region-Based Description、Angular Radial Transform、Contour-Based Description）to be a multi-descriptions retrieve interface. In our retrieve system, user can using the original image with shape features to extract the trademark or logo from the image database. We also propose a drawing panel. For complicated shape, user cans just drawing a simple shape, then the system will extract the family images. For the image database classify, we divided them into two parts. One is fully and closed contour image database. The other is complicated and multi-regions image database. According the image features, user can choice the database to extract the image. This issue proposed a easily and humanity operated interface.

摘要I
ABSTRACTIII
誌謝V
圖目錄VIII
表目錄XI
第一章 導論1
1.1 研究動機與研究目的1
1.2 論文方法與貢獻3
1.3 論文結構5
第二章 影像檢索系統簡述6
2.1 徑角轉換8
2.2 曲率尺度空間11
2.3 以影像內容物件外形輪廓為主的描述方法14
2.4 以影像內容物件外形區域為主的描述方法16
2.5 以距離角度分佈為基礎的分層外形描述方法18
第三章 影像資料庫分類與建置23
3.1 完整且封閉輪廓影像資料庫26
3.2 複雜且多區域影像資料庫27
第四章 影像檢索介面之建構31
4.1 多種的查詢（QUERY）介面選擇方式33
4.2 不同特徵的選擇方式37
4.3 不同的資料庫選擇方式38
4.4 漸進式檢索修正39
第五章 實驗結果41
第六章 結論與未來研究方向52
6.1 結論52
6.2 未來研究方向53
參考文獻54

中文部份
[1]黃丁順,’應用於商標及圖示檢索的分層外形特徵描述子’,義守大學碩士論文,2004。
英文部份
[1]A. K. Jain and A.Vailaya, “Shape-based retrieval: a case study with trademark image databases,” Pattern Recognition, Vol. 31, No. 9, pp.1369-1390, 1998.
[2]A. Yamada, M. Pickering, S. Jeannin and L. C. Jens, ” MPEG-7 Visual part of Experimentation Model Version 9.0-Part 5 Region-Based Shape,” ISO/IEC JTC1/SC29/WG11/N3914, January 2001.
[3]A. Yamada, M. Pickering, S. Jeannin and L. C. Jens, ” MPEG-7 Visual Partof Experimentation Model Version 9.0-Part 5 Contour -Based Shape,” ISO/IEC JTC1/SC29/WG11/N3914 Pisa, January 2001
[4]Del Bimbo, P. Pala, “Visual image retrieval by elastic matching of user sketches,” IEEE Trans, Pattern Anal. Mach. Intelligence, Vol. 19, February 1997
[5]G. Woods, ”Digital Image Processing,” Addison-Wiley Publishing Company, Inc. U.S. 1992.
[6]H. K. Kim and J. D. Kim, “Region-base shape descriptor invariant to rotation scale and translation,” Signal Processing: Image Communication 11.2000, pp.87-93.
[7]Hui Xu, Mengyang Liao, “Cluster-Based Texture Matching for Image Retrieval,” Image Procseeing ICIP 98. Proceedings International Conference on, Vol. 2, pp. 766-769,1998.
[8]M. Bober , ”MPEG-7 visual shape descriptors,” IEEE Transaction on Circuit and System for Video Technology, vol.11, NO.6 June 2001, pp.716-719.
[9]MY. Ma and BS Manjunath, “NeTra: a Toolbox for Navigating Large Image Databases,”Multimedia Systems, Vol. 7, pp 184-198,1999.
[10]QBIC Project, IBM Research, http://wwwqbic.almaden.ibm.com
[11]R. Distasi, D. Vitulano and S. Vttulano, ”A Hierarchical Representation for Content-based Image Retrieval,” Journal of visual languages and computing, 11.2000, pp.369-382.
[12]Smith, J.R., and Chang, S.F., “VisualSEEk: a fully automated content-based image query system,” in Proceedings ACM Multimedia, Boston MA, ACM Press, 87-98, 1996.
[13]Th. Gevers and AWM Smeulders, “PicToSeek: Combining Color and Shape Invariant Features for Image Retrieval pdf file,” IEEE Trans. on Image Processing, Vol. 9, No. 1,pp. 102-119,January, 2000.
[14]W. Y. Kim and Y. S. Kim, “A region-based shape descriptor using Zernike Moments,” Signal Processing: Image Communication, pp.95–102, 2000.